[1]
Zhu, L. et al. 2024. SEER: Backdoor Detection for Vision-Language Models through Searching Target Text and Image Trigger Jointly. Proceedings of the AAAI Conference on Artificial Intelligence. 38, 7 (Mar. 2024), 7766–7774. DOI:https://doi.org/10.1609/aaai.v38i7.28611.